I have been trying to follow a more disciplined approach to branching in my
projects to allow easier tracking and regression. For example, I learnt that
it's a good idea to prefix branch names with category tokens, such as
feature/
, bug/
, or chore/
. This allows us to use wildcards when listing
branches:
$> git branch --list bug/*
$> git branch --list feature/*
feature/X49972617
Since I use Pivotal Tracker, I name my branches <type>/X<story-id>
, and then
use a simple hook to prepend all of my commit messages with the story ID.
$> git commit
[#49972617]
# Please enter the commit message for your changes. Lines starting
# with '#' will be ignored, and an empty message aborts the commit.
# On branch feature/X49972617